In a PAC equation, the error correction term must be written as:
β×(yₜ₋₁-xₜ₋₁)
where x is the endogenous variable determined by the equation and y
the target for x. The ordering matters, if not respected the
preprocessor wll not identify the error correction term (hence the
target) and will throw an error. The target must be an endogenous
variable or the log of an endogenous variable. For a more general
target, ie an expression, one can create an auxiliary variable.
We impose the ordering, target comes first in the error correction
term, so that the stability condition is always β>0 (this constraint
can be enforced when estimating the PAC equation).
Those tests broke because they were declaring an endogenous (var_allows_…) that
is not used in the model, and this is no longer allowed by the preprocessor.
Those variables were introduced by commit
dccabecee7 (on the old ecb-master branch), at the
same time as a second PAC equation was introduced.
The fix consists in making the second PAC equation different from the first
one, defining a different endogenous.
Added the possibility to associate more than one trend to an endogenous
variable. The number of error correction equations is longer required to be
equal to the number of trends.